Devices for Continuous Glucose Monitoring have advanced over the past decade, but there is not yet a glucose monitoring system that provides the precision necessary for automated control of insulin delivery in diabetic patients. In particular, protecting patients against hypoglycemia could allow more aggressive insulin therapy and a closed-loop glucose control system could be enabled. Here we describe a novel biosensor Continuous Glucose Monitoring concept based on fluorescent protein engineering and exploiting glucose- responsive protein domains that already exist in nature. The ultimate goal of this project is to develop a Continuous Glucose Monitoring cell-based biosensor that can wirelessly transmit to the patient's insulin pump and control insulin delivery all day. There are three critical components of this Continuous Glucose Monitoring concept, and thus this project requires an interdisciplinary team working in concert. The first Aim of this project is to generate a high signal-to-noise genetically encoded fluorescence intensity sensor for glucose that will be suitable for the in vivo monitoring technology. We also need a device that can interpret the fluorescent intensity change in the skin and transmit to the insulin pump. Thus, the second Aim of this project is to develop and optimize a fluorescence measurement device for detection of the fluorescence change. We have already initiated design of this device, which we predict could be manufactured for a low price. Finally, we need to introduce the protein into the patient's skin, so that the new device can record the fluorescence change. Thus, the third Aim is to explore both gene transfer and autologous cell transfer experiments in order to deliver the fluorescent biosensor protein in a safe and practical manner. This is a critical and challenging goal, as successful translation will require that the system be stable and devoid of genotoxicity. Although our Continuous Glucose Monitoring concept is high risk and faces major challenges, a solution to the Continuous Glucose Monitoring problem would be of great clinical significance. Redundant monitoring systems may be necessary for optimal glucose control, and independent technologies may reduce the noise inherent in all of these monitoring technologies. For example, a cell-based system as described here could interface with a non-biological spectroscopic system to provide redundant signals for preventing hypoglycemia. Ultimately, a combined modality Continuous Glucose Monitoring biosensor in conjunction with an insulin pump could allow reliable closed-loop glucose control.